Abstract
Intelligent wheelchairs aim to improve mobility limitations by providing ingenious mechanisms to control and move the chair. This paper aims to enhance the autonomy level of intelligent wheelchair navigation by applying reinforcement learning algorithms to move the chair to the desired location. Also, as a second objective, add one more chair and move both chairs in pairs to promote group social activities. The experimental setup is based on a simulated environment using gazebo and ROS where a leader chair moves towards a goal, and the follower chair should navigate near the leader chair. The collected metrics (time to complete the task and the trajectories of the chairs) demonstrated that Deep Q-Network (DQN) achieved better results than the Q-Learning algorithm by being the unique algorithm to accomplish the pair navigation behaviour between two chairs.

Abstract
Robotics is an ever-growing field, used in countless applications, from domestic to industrial, and taught in advanced courses of multiple higher education institutions. Robot Operating System (ROS), the most prominent robotics architecture, integrates several of these, and has recently moved to a new iteration in the form of ROS2. This project aims to design a complete educational package meant for teaching intelligent robotics in ROS1 and ROS2. A foundation for the package was constructed, using a small differential drive robot equipped with camera-based virtual sensors, a representation in the Flatland simulator, and introductory lessons to both ROS versions and Reinforcement Learning (RL) in robotics. To evaluate the package's pertinence, expected learning outcomes were set and the lessons were tested with users from varying backgrounds and levels of robotics experience. Encouraging results were obtained, especially in the ROS1 and ROS2 lessons, while the feedback from the RL lesson provided clear indications for future improvements. Therefore, this work provides solid groundwork for a more comprehensive educational package on robotics and ROS.

Abstract
To tackle wildfires and improve forest biomass management, cost effective and reliable mowing and pruning robots are required. However, the development of visual perception systems for forestry robotics needs to be researched and explored to achieve safe solutions. This paper presents two main contributions: an annotated dataset and a benchmark between edge-computing hardware and deep learning models. The dataset is composed by nearly 5,400 annotated images. This dataset enabled to train nine object detectors: four SSD MobileNets, one EfficientDet, three YOLO-based detectors and YOLOR. These detectors were deployed and tested on three edge-computing hardware (TPU, CPU and GPU), and evaluated in terms of detection precision and inference time. The results showed that YOLOR was the best trunk detector achieving nearly 90% F1 score and an inference average time of 13.7ms on GPU. This work will favour the development of advanced vision perception systems for robotics in forestry operations.

Abstract
Bin picking is a challenging problem that involves using a robotic manipulator to remove, one-by-one, a set of objects randomly stacked in a container. When the objects are prone to entanglement, having an estimation of their pose and shape is highly valuable for more reliable grasp and motion planning. This paper focuses on modeling entangled tubes with varying degrees of curvature. An unconventional machine learning technique, Inductive Logic Programming (ILP), is used to construct sets of rules (theories) capable of modeling multiple tubes when given the cylinders that constitute them. Datasets of entangled tubes are created via simulation in Gazebo. Experiments using Aleph and SWI-Prolog illustrate how ILP can build explainable theories with a high performance, using a relatively small dataset and low amount of time for training. Therefore, this work serves as a proof-of-concept that ILP is a valuable method to acquire knowledge and validate heuristics for pose and shape estimation in complex bin picking scenarios.

Abstract
The manipulation of deformable objects poses a significant challenge for the automotive industry. In particular, the assembly of flexible cables and wire-harnesses in vehicles is still performed manually as there is yet to be a reliable and general solution for this problem. This paper presents a simple yet efficient motion planning algorithm to mount a flexible wire in an assembly jig, where the wire must traverse a set of forks in order. The algorithm uses a heuristic based on a set of control points to guide the wire's movement. Various controlled assembly scenarios are built in simulation using MuJoCo, a physics engine that can emulate the dynamics of Deformable Linear Objects (DLO). Experimental results in simulation demonstrated that the amount and orientation of the forks has a large impact in the solution's performance and highlighted several key ideas and challenges moving forward. Thus, this work serves as a stepping stone towards the development of more complete solutions, capable of assembling flexible items in vehicles.

Abstract